ECE-470-Final-Project

Team Name: Book Worm

Team Members: Ali Albazroun - aia

The goal of this project is to simulate a robot that could sort books by size and color using CoppeliaSim and Python.

Project YouTube playlist: https://www.youtube.com/playlist?list=PLpN81Jj_B-xPhZFfH4yKj61GJNcMmq5qz

Project Update 1:

Got UR3 robot arm to move to a specified position and back, and got a Spherical Vision Sensor to output depth and color data.

• Used simxSetJointTargetPosition to move the arm.

• Used simxGetVisionSensorDepthBuffer and simxGetVisionSensorImage to get depth and color data respectively. Thanks to "Am I two?"\n"No, UR3." Team for their python code which helped me understand how to use the commands to interface with CoppeliaSim.

Project Update 2:

Got the UR3 robot to respond to a book passing by on a conveyor belt and created forward kinematic model for UR3 robot arm.

• Defined a function to calculate the pose of the end effector from given joint angles using forward kinematics.

• Proximity sensor detects a book on conveyor belt and sends a signal using simxsetFloatSignal.

• Signal is received in python using simxgetFloatSignal then the robot arm does a 180º base rotation.

Project Update 3:

Got the Color Sorter UR3 arm to color sort book passing by the conveyor belt using inverse kinematics.

• Defined a function to calculate the joint angles of a robot given the world coordinates of the gripper.

• Created an algorithm to sort the books using vision sensor data from simxGetVisionSensorImage.

• Set up communication between proximity sensors, conveyor belts, and UR3 arms using simxgetFloatSignal and simxsetFloatSignal.

Final Project Update:

Got the Size Sorter UR3 arm to sort books by size using inverse kinematics.

• Created an algorithm to sort the books using data from two proximity sensors.

• Redefined the moveJoints function to allow the robots to pick and place the books without knocking the rest down.

• Refined the communication between proximity sensors, conveyor belts, and UR3 arms.

• Fine tuned inverse kinematics for better stacking.